ArticlePDF Available

An Assessment of Outcomes in Outdoor Behavioral Healthcare Treatment

Authors:

Abstract and Figures

Outdoor behavioral healthcare (OBH) is an emerging treatment that utilizes wilderness therapy to help adolescents struggling with behavioral and emotional problems. The approach involves immersion in wilderness or comparable lands, group living with wilderness leaders and peers, and individual and group therapy sessions facilitated by licensed therapists in the field. OBH also offers educational and psychoeducational curriculum all designed to reveal and address problem behaviors, foster personal and social responsibility, and enhance the emotional growth of clients. The extant studies on the effectiveness of OBH and wilderness therapy reveal consistent lack of theoretical basis, methodological shortcomings and results that are difficult to replicate. This publication reports the results of an outcome assessment for adolescent clients who received treatment in seven participating OBH programs that averaged 45 days in length from May 1, 2000 to December 1, 2000. Adolescent client well-being was evaluated utilizing the Youth Outcome Questionnaire (Y-OQ) and the Self Report-Youth Outcome Questionnaire (SR Y-OQ) (Burlingame, Wells, & Lambert, 1995). Complete data sets at admission and discharge were collected for 523 client self-report and 372 parent assessments. Results indicated that at admission clients exhibited presenting symptoms similar to inpatient samples, which were on average significantly reduced at discharge. Follow-up assessments using a random sample of clients found that on average, outcomes had been maintained at 12-months posttreatment.
Content may be subject to copyright.
Idaho Forest, Wildlife, and
Range Experiment Station
Moscow, Idaho
Director
Steven Daley Laursen
TT
TT
Tecec
ecec
echnical Rhnical R
hnical Rhnical R
hnical Ree
ee
eporpor
porpor
port 28t 28
t 28t 28
t 28
August 2002August 2002
August 2002August 2002
August 2002
KK
KK
Keith Ceith C
eith Ceith C
eith C..
..
. R R
R R
Russell,ussell,
ussell,ussell,
ussell, Ph.D Ph.D
Ph.D Ph.D
Ph.D..
..
.
Outdoor Behavioral Healthcare Research Cooperative
in the
University of Idaho-Wilderness Research Center
Longitudinal Assessment of
Treatment Outcomes in
Outdoor Behavioral
Healthcare
August 2002
University of
Idaho
Ascent, Naples, Idaho
Anasazi Foundation, Mesa, Arizona
Aspen Achievement Academy, Loa, Utah
Catherine Freer Wilderness Therapy, Albany, Oregon
Redcliff Ascent, Springville, Utah
Sunhawk Academy, St. George, Utah
SUWS, Shoshone, Idaho
Three Springs, Huntsville, Alabama
Trailhead Wilderness School, Georgetown, CO
THE AUTHOR
Keith C. Russell, Leader, Outdoor Behavioral Healthcare Research Cooperative and assistant research professor
in the Department of Resource Recreation and Tourism, University of Idaho, Moscow, Idaho 83844-1144, E-mail:
keithr@uidaho.edu; Website: www.its.uidaho.edu/wrc
ACKNOWLEDGMENTS
The technical reviewers of this publication are gratefully acknowledged. While their comments, suggestions, and
insights helped us greatly improve the document, they bear no responsibility for the final presentation. Dr. John
Hendee, Director of the UI-Wilderness Research Center, is also acknowledged for his support and commitment to
this research project and for his reviews of the study.
Peer Reviewers
Rob Cooley, Ph.D., Cooley and Associates and Catherine Freer Wilderness Therapy Expeditions, Albany,
Oregon
Thomas J. Doherty, Psy.D., Clinical Supervisor, Catherine Freer Wilderness Therapy Expeditions, Albany,
Oregon
Alan Ewert, Ph.D., Professor, Patricia and Joel Meier Outdoor Leadership Chair, Department of Parks and
Recreation Administration, Indiana University, Bloomington, Indiana
Steven Hollenhorst, Ph.D., Department Chair, Department of Resource Recreation and Tourism, University
of Idaho, Moscow, ID
Gil Hallows, Director, Aspen Achievement Academy, Loa, Utah
Mike Merchant, Chief Operating Officer, Anasazi Foundation, Mesa, Arizona
We also acknowledge the financial support of member programs of the Outdoor Behavioral Healthcare Industry
Council (OBHIC), and the University of Idaho-Wilderness Research Center in the College of Natural Resources,
whose support and cooperation were vital to the research.
OBHIC Member Programs This publication is issued as contribution No. 945
of the Idaho Forest, Wildlife and Range Experi-
ment Station, College of Natural Resources,
University of Idaho, Moscow, ID 83844-1144.
To enrich education through diversity, the
University of Idaho is an equal
opportunity/affirmative action employer.
Copyright © 2002 by the University of Idaho-
Wilderness Research Center. All rights
reserved.
2
EXECUTIVE SUMMARY
This publication reports the results of a follow-up assessment of treatment outcomes for adolescent
clients who received treatment in seven participating Outdoor Behavioral Healthcare (OBH) programs
that averaged 45-days in length from May 1, 2000 to December 1, 2000. Adolescent client well-being
was evaluated utilizing the Youth Outcome Questionnaire (Y-OQ) (Burlingame, Wells, & Lambert,
1995). The initial assessment of treatment outcomes reported in Technical Report 27 (Russell, 2001)
showed that clients had made significant improvement from treatment to discharge as measured by
the Y-OQ. High Y-OQ scores (i.e. greater than 85) indicate severe problems in adolescent’s lives, and
low scores indicate a normal range of functioning (i.e. 46 or lower). Client self-report mean Y-OQ
scores were 70.67 at admission and 47.55 at discharge, indicating an average score reduction of more
than 20 points. Parent assessment Y-OQ mean scores were 101.19 at admission and 48.55 at dis-
charge, indicating an average reduction of 52.64. Discharge scores for both client self report and
parent assessment were close to the normal range of symptoms (46 or below) as established by
Burlingame et al. (1995b) in their sample tests of normal populations.
An important question asked in the conclusion of this report was: To what degree will clients
maintain outcomes realized from OBH treatment at follow-up periods? This question is especially
important given the lack of longitudinal outcome studies found in the literature and the uncertainty
surrounding the degree to which clients can apply the skills and lessons learned in OBH treatment to
their daily lives (Russell, 2001). The results indicated that at the 3- and 6-month follow-up periods,
outcomes were maintained as indicated by client self-report scores that did not significantly differ
from previous assessments; parent assessments indicated higher Y-OQ scores at 3-month (4 points)
and 6-month (8-points) follow-up periods when compared to discharge scores, suggesting a deteriora-
tion of outcomes (score differences were not statistically significant). It was assumed based on
supporting literature and statistically significant score differences between full and incomplete data
sets that these samples could be biased in the direction of more favorable outcomes. This led to a
random sample of clients to be contacted at the 12-month follow-up period to gain a more representa-
tive and unbiased sample.
For the randomly sampled data set at 12-months (scores at admission, discharge, and 12-months),
clients self reported outcomes that averaged 8 points under the cut-score of 46 (38.61), demonstrating
that those clients had maintained outcomes from treatment, and had actually continued to improve up
to one-year after completion of treatment. Parent scores were almost 10 points higher than client self-
reports at 48.67, but were also close to the cut-score of 46 points. These findings suggest that clients
were doing well emotionally and behaviorally at the 12-month follow-up period.
No statistical differences were found in average scores at admission, discharge, and 12-month follow-
up when comparing clients who utilized aftercare services with those who returned home. Clients self-
reported significant improvement in the behavioral dysfunction content area of the Y-OQ, suggesting
improvement in organizing tasks, completing assignments in school, and learning how to handle
frustration in appropriate ways. Both parents and clients also noted a deterioration in the interper-
sonal relations content area of the Y-OQ, despite a reported significant improvement in this area
found at discharge from treatment. This assesses clients relationship with parents, and other adults,
as well as interaction with friends, aggressiveness, arguing and defiance. These results suggest that
OBH treatment can be effective at addressing presenting behavioral problems, but may need to
further identify ways to help clients maintain recently developed interpersonal skills that are continu-
ally tested in post treatment environments.
3
TABLE OF CONTENTS
ACKNOWLEDGMENTS ................................................................................................. 2
EXECUTIVE SUMMARY ........................................................................................................... 3
TABLE OF CONTENTS ........................................................................................................... 4
LIST OF TABLES ................................................................................................................... 6
LIST OF FIGURES .................................................................................................................. 7
INTRODUCTION ...................................................................................................................... 8
RESEARCH METHODS ......................................................................................................... 10
Research Design ........................................................................................................ 10
Research Questions ................................................................................................... 10
Youth Outcome Questionnaire (Y-OQ) ......................................................................11
Criteria to Assess the Y-OQ and SR Y-OQ ............................................................. 12
Clinically Meaningful Change ..................................................................................... 13
Data Collection and Entry .......................................................................................... 14
Study Retention .......................................................................................................... 16
12-Month Follow-up Assessment Procedures ........................................................... 17
Data Analysis ............................................................................................................. 18
LIMITATIONS TO THE STUDY................................................................................................19
Interpreting Follow-up Results ................................................................................... 21
Summary of Limitations ............................................................................................. 21
4
RESULTS ............................................................................................................................ 22
Client Characteristics ................................................................................................. 22
Gender and Age .......................................................................................................22
Primary Diagnoses...................................................................................................23
Prior Treatment .........................................................................................................24
Summary of OBH Client Characteristics................................................................... 24
TREATMENT OUTCOME .......................................................................................................25
Research Question 1. What outcomes that resulted from OBH treatment
were still present at 3-, 6-, and 12-month follow-up periods? ................... 25
Outcomes at 3-month Follow-up Period ..........................................................26
Outcomes at 6-month Follow-up Period ..........................................................27
Outcomes at 12-month Follow-up Period ........................................................29
Research Question 2. How did OBH clients’ emotional and behavioral
functioning vary according to client attributes of age and gender?......... 31
Treatment Outcome by Gender at 12-Month Follow-up................................. 31
Treatment Outcome by Age at 12-Month Follow-up .......................................32
Research Question 3. How did OBH clients’ emotional and behavioral
functioning vary according to utilization of aftercare services?............. 33
Research Question 4. How did OBH clients’ emotional and behavioral
functioning vary across the six content areas of the Y-OQ?................... 36
DISCUSSION ........................................................................................................................37
SUMMARY AND CONCLUSIONS ..............................................................................................40
LITERATURE CITED ............................................................................................................42
5
LIST OF TABLES
Table 1. Participant programs, program type, model and length and program time
spent on wilderness expedition (Russell & Hendee 2000). ................................................ 11
Table 2. Total study participants, number who completed at least one Y-OQ
during study period, sample size required to estimate the average scores at 12-months,
total number of study participants contacted and total number of 12-month
follow-up contact calls. ..................................................................................................... 17
Table 3. Parent and Y-OQ scores at admission, discharge, 3-, 6-, and 12-months
arranged by total assessment for that time period, and in 3-month, 6-month, and
12-month data sets. ........................................................................................................... 19
Table 4. Total number of clients participating in OBH treatment and participating
in the study from May 1, 2000 to December 1, 2000 in the eight
participating programs. ...................................................................................................... 20
Table 5. Gender of study participants. ..................................................................................... 22
Table 6. Age of study participants. ......................................................................................... 22
Table 7. Average percentage of diagnoses of study participants............................................ 23
Table 8. Frequency and percentage of clients who received inpatient and outpatient
treatment services prior to enrolling in OBH treatment. ..................................................... 24
Table 9. Admission and discharge average SR Y-OQ and Y-OQ scores, including
mean difference in scores from admission to discharge. ................................................... 25
Table 10. Average SR Y-OQ and Y-OQ scores for the incomplete data set
and the 3-month data set. .................................................................................................. 26
Table 11. Average SR Y-OQ and Y-OQ scores for the incomplete data set and
the 6-month data set. ......................................................................................................... 27
Table 12. Average SR Y-OQ and Y-OQ scores for the incomplete data set and
the 12-month data set........................................................................................................ 29
Table 13. Average SR Y-OQ and Y-OQ scores for the 12-month data set
for male and female study participants. ............................................................................. 31
Table 14. Six Average SR Y-OQ and Y-OQ scores and frequency for the
12-month data set for clients classified into three age groups: ages 13-14,
ages 15-16, ages 17-19. ...................................................................................................... 32
Table 15. Six Average SR Y-OQ and Y-OQ scores and frequency for the
12-month data set for clients classified into three age groups: ages 13-14,
ages 15-16, ages 17-19. ...................................................................................................... 32
Table 16. Average Y-OQ scores at admission, discharge and 12-months for
all clients who utilized aftercare services and those who returned home.. ........................ 34
Table 17. Average Y-OQ score for complete data sets at admission, discharge
and 12-months for 12-month data set who utilized aftercare services
and those who returned home.. ......................................................................................... 35
Table 18. Average Y-OQ subscale scores for discharge and 12-months,
including cutscores established by Burlingame et al. (1996) for each of
the six content areas. ......................................................................................................... 36
6
LIST OF FIGURES
Figure 1. Six content areas of the Youth-Outcome Questionnaire (Y-OQ). ............................. 12
Figure 2. Boxplot depicting the range and median of client self-report Y-OQ scores at
admission, discharge, 3-, and 6-months. ............................................................................ 28
Figure 3. Boxplot depicting the range and median of parent assessed Y-OQ scores at
admission and 12-months. ................................................................................................. 28
Figure 4. Boxplot depicting the range of parent assessed Y-OQ scores at
admission, discharge, 3-, and 6-months. ............................................................................ 30
7
8
INTRODUCTION
This publication reports the results of a follow-up assessment of treatment
outcomes for adolescent clients who received treatment in one of seven partici-
pating Outdoor Behavioral Healthcare (OBH) programs from May 1, 2000 to
December 1, 2000. Clients and their parents were contacted at 3-, 6-, and 12-
month follow-up periods to evaluate their well-being utilizing the Youth Out-
come Questionnaire (Y-OQ) (Burlingame, Wells, & Lambert, 1995). The initial
assessment of treatment outcomes reported in Technical Report 27 (Russell,
2001) showed that clients had made significant improvement from treatment to
discharge as measured by the Y-OQ. Client self-report mean Y-OQ scores
were 70.67 at admission and 47.55 at discharge, indicating an average score
reduction of more than 20 points. Parent assessment Y-OQ mean scores were
101.19 at admission and 48.55 at discharge, indicating an average reduction of
52.64. Thus, parents rated the clients presenting symptoms as more severe
than did the clients themselves, but they perceived symptoms at discharge that
were very similar. Discharge scores for both client self report and parent
assessment are close to the normal range of symptoms (46 or below) as estab-
lished by Burlingame et al. (1995b) in their sample tests of normal populations.
Results of this study indicated that participation in OBH programs led to a statisti-
cally significant reduction in the severity of behavioral and emotional symptoms, as
perceived by the clients, and even more so by their parents, as measured by the Y-
OQ questionnaire. These findings suggest that OBH could be effective at
helping reduce the behavioral, emotional, and psychological symptoms with
which clients presented. An important question asked in the conclusion of this
report was: To what degree will clients maintain outcomes realized from
OBH treatment at follow-up periods? This question is especially important
given the lack of longitudinal outcome studies found in the literature and the
uncertainty surrounding the degree to which clients can apply the skills and
lessons learned in OBH treatment to their daily lives (Russell, 2001).
These positive findings were consistent with several meta-analyses that suggest
that OBH and related wilderness programs for troubled adolescents enhance
self concept (Hattie, Marsh, Neill, & Richards, 1997), strengthen locus of
control (Hans, 2000), and help reduce recidivism rates of young offenders
(Winterdyk & Griffiths, 1984). However, most participant assessments are
done immediately following the experience and could possibly be affected by
the “post-group euphoria,” a feeling of euphoria often experienced by the
participant following an intense and unique group experience (Ewert &
Heywood, 1991; Marsh, Richards, & Barnes, 1986). Follow-up assessments of
treatment outcomes can address this concern and are especially important in
OBH because of the primary goal of treatment: helping adolescent clients
understand what issues in their lives may be driving their problem behaviors,
and creating a desire to want to change these behaviors (see Russell &
Hendee, 2000, for overview of OBH treatment). Moreover, the degree to
which clients can apply skills and lessons learned in wilderness environments to
their everyday lives has not been well documented in the literature (Hattie et
al., 1997; Winterdyk & Griffiths, 1984). Finally, very few long term assess-
ments of client outcomes in OBH related programs have been conducted.
An important question
asked in the conclu-
sion was: To what
degree will clients
maintain outcomes
realized from outdoor
behavioral healthcare
treatment at follow-
up periods?
9
The few longitudinal studies found in the literature have typically examined
recidivism rates for delinquent and substance abusing youth. Hattie et al.
(1997) found in a meta-analysis of different participants in OBH related pro-
grams that “the effect sizes for the delinquents in the follow-up studies were
greater than for the other identified groups” (p. 59). This meta-analysis in-
cluded some of the seminal studies on Outward Bound (Kelly & Baer, 1968,
1969; Kelly, 1974) and highlighted the potential of wilderness expeditions to help
rehabilitate young offenders not being reached by traditional corrections pro-
grams. Castellano and Sodersrom (1992) continued this line of research, and
found that after taking 30 adolescents on a 30-day wilderness expedition,
successful completion of the course resulted in arrest reductions which began
immediately after the program was complete and lasted for about one year.
Another study found that a 3-day therapeutic camping program for young
people as part of a substance abuse treatment program had a positive impact on
relapse rates for participants when compared to controls at a 10-month follow-
up period (Bennet, Cardone, & Jarczyk, 1998). Despite these positive results,
the question still remains: Do OBH programs do a better job of reducing
recidivism for adolescents in these programs than traditional treatment and
rehabilitation programs?
For example, several longitudinal studies have failed to conclude significant
differences in recidivism rates between OBH programs and other methods of
rehabilitation. Deschenes and Greenwood (1998) evaluated the Nokimus
Challenge Program for delinquents, which utilizes a wilderness expedition as a
component of rehabilitation, and found few differences in outcomes of recidi-
vism and social adjustment measures between treatment and control groups at
follow-up periods. The authors suggest that to derive benefit from short-term
placements, the aftercare component must be strengthened to help youth avoid
relapse. These findings are echoed by Eggleston (1998), who examined adjudi-
cated youth aged 13-18, who participated in an OBH program in New Zealand.
She interviewed participants 18-months after the program to examine which
program elements were still important in their lives. She concluded that pro-
gram benefits were difficult for participants to apply in their everyday lives, and
that follow-up care was inadequate. Each of these studies report either no
differences between OBH treatment and control groups, or diminished effects
of treatment due to lack of follow-up by programs and point to the need for
aftercare services as essential for participants to maintain outcomes. These
findings are supported by the extensive literature on the effectiveness of sub-
stance abuse treatment programs for adolescents that demonstrates the impor-
tance of aftercare and relapse prevention in maintaining outcomes (Winters,
1999).
Longitudinal outcome
studies have identified
diminished effects over
time, highlighting the
importance of aftercare
services to help clients
maintain outcomes
after treatment.
10
In summary, any examination of OBH outcomes at follow-up periods must take
into account how aftercare services are utilized because of the important role
they play in maintaining treatment outcomes. As Hattie et al. (1997) state
“recidivism may be, at least in part, a function of inadequacy of postprogram
support” (p. 59). This is supported by OBH program staff who believe that
short-term (3-8 weeks) OBH programs are designed to help prepare and
develop an internal motivation for clients to engage in long term care when
appropriate (Russell, 1999). Second, those studies that have assessed out-
comes at follow-up periods have found mixed results. One reason that has
been suggested, and that is consistent with similar literature on the treatment of
at-risk adolescents in other modalities, is the presence of aftercare and follow-
up procedures to help explain these mixed results. Reflecting this critical need
to assess outcomes at follow-up periods and examine how aftercare services
are utilized, this study will report results from 3-, 6-, and 12-month assessments
of OBH clients that either returned home, or went on to an aftercare program
at the termination of treatment.
RESEARCH METHODS
Research Design
A time series research design was used in this study (Graziano & Raulin,
1997). A census of 858 clients at seven participating programs (see Table 1)
were surveyed during the time period of May 1, 2000 to December 1, 2000.
The seven participating programs are all members of the Outdoor Behavioral
Healthcare Industry Council (OBHIC) but do not represent the entire OBH
industry. Therefore, study findings cannot be generalized beyond these seven
programs; findings represent possible outcomes under the parameters of treat-
ment in these programs.
Research Questions
The following research questions were addressed in this study:
Research Question 1. What outcomes that resulted from OBH treatment as
measured by Y-OQ score differences between admission and discharge were
still present at 3-, 6-, and 12-month follow-up periods?
Research Question 2. How did former OBH clients’ emotional and behav-
ioral functioning at follow-up periods vary according to client attributes, such as
age and gender?
Research Question 3. How did former OBH clients’ emotional and behav-
ioral functioning at follow-up periods vary according to the utilization of after-
care services?
Research Question 4. How did former OBH clients’ emotional and behav-
ioral functioning at follow-up periods vary across the six content areas of the
Y-OQ?
11
Table 1. Participant programs, program type, model and length and program time spent
on wilderness expedition (Russell & Hendee 2000).
Youth Outcome Questionnaire (Y-OQ)
The Outcome Questionnaire (OQ) was developed to assess outcomes in behav-
ioral healthcare and has been deemed a valid and reliable psychometric tool
(Lambert, Huefner, & Reisenger, 1996). The Youth Outcome Questionnaire
(Y-OQ™) and Self-Report-Youth Outcome Questionnaire (SR Y-OQ™)
evolved from the OQ, (herein referred to simply as the Y-OQ except where
distinction is important), and offers parent assessment and adolescent self-
reports of an adolescents’ well-being and is designed for repeated measure-
ment of client symptoms (i.e. admission, during therapy, at termination, and at
follow-up intervals) (Burlingame et al., 1995b; Burlingame et al., 1996; Lambert
& Cattani-Thompson, 1996; Lambert et al., 1992; Lambert et al., 1996; Wells,
Burlingame, Lambert, Hoag, & Hope, 1996; Wells, 1990).
Organization Type Model Length Time Spent on
Wilderness Expe dition
Anasazi
1424 S. Stapley
Mesa, Arizona 85204
Private
Placement Continuous Flow
Expedition 56 days 56 days
Ascent
PO Box 230
Ruby Creek Road
Naples, Idaho 83847
Private
Placement Base Camp
Expedition 42 days 14 days
Aspen Achievement Academy
PO Box 369
Loa, Utah 84747
Private
Placement Continuous Flow
Expedition 52 days 52 days
Catherine Freer
PO Box 1064
Albany, Oregon 97321
Private
Placement Contained
Expedition 21 days 21 days
Red Cliff Ascent
757 S. Main Street
Springvale, Utah 84663
Private
Placement Continuous Flow
Expedition 56 days 56 days
SunHawk Academy
948 N 1300 W
St. George, Utah 84770
Private
Placement Residential
Expedition 56 days 28 days
SUWS
911 Preacher Creek Road
Shoshone, Idaho 83352
Private
Placement Contained
Expedition 21 days 21 days
Ave. 45 days Ave. 34.5 days
12
A total of 64 questions are included in the Y-OQ to assess change in the six
content areas described in Figure 1.
Figure 1. Six content areas of the Youth Outcome Questionnaire (Y-OQ).
The Y-OQ parent assessment is designed to measure parent perceptions of a
wide range of behaviors, situations, and moods which commonly apply to
troubled teenagers. The Y-OQ self-report is designed to measure client self
assessments of the same behaviors. When the client is admitted to treatment,
the Y-OQ is completed by parents and the client to establish baseline scores
against which to compare future scores. Follow-up periods in this longitudinal
study were at 3-, 6-, and 12-months. The 64 items contained in the Y-OQ are
summed across the six content areas to produce a total score.
Criteria to Assess the Y-OQ and SR Y-OQ
The Y-OQ assesses the psychological, symptomatic and social functioning of
adolescents, which reflect the goals of OBH treatment. It is also a well-normed
and easily administered outcome measure with good internal consistency and test-
re-test reliability. As a general rule, validity and reliability coefficients should be at
or above .80 (Jacobsen & Truax, 1991). Estimates of the Y-OQ internal consis-
tency range from .74 to .93 with a total scale estimate of .96. Test-re-test reliabil-
ity scores are above .70, indicating moderately high temporal stability (see
Burlingame et al., 1996 for detailed review of these estimates). High correlations
also exist between the Y-OQ and subscale scores, and other frequently used
assessment instruments (Wells et al., 1996). For example, scales on the Child
Behavior Checklist (Achenbach, 1991) correlate highly with parallel scales on the
Y-OQ. The Y-OQ instrument can be easily administered by staff at each OBH
program and only takes ten minutes for the parents and client to complete. The
device has not proven too complicated or detailed for respondents, which is an
important consideration when working with adolescents (Burlingame et al., 1996).
Content Area Assesses
Intrapersonal Distress (ID) Assesses change in emotional distress including anxiety, depression, fearfulness,
hopelessness, and self harm.
Somatic (S) Assesses change in somatic distress typical in psychiatric presentation, including
headaches, dizziness, stomachaches, nausea, and pain or weakness in joints.
Interpersonal Relations (IR) Assesses change in the child's relationship with parents, other adults, and peers as well as
the attitude towards others, interaction with friends, aggressiveness, arguing, and defiance.
Critical Ite ms (C I) Assesses inpatient services where short term stabilization is the primary change sought:
changes in paranoia, obsessive-compulsive behavior, hallucination, delusions, suicide,
mania, and eating disorder issues.
Social Problems (SP) Assesses changes in problematic behaviors that are socially related, including truancy,
sexual problems, running away from home, destruction of property and substance abuse.
Behavioral Dysfunction (BD) As sesses c hange in a child's a bility to organize tasks, comp lete ass ignme nts, c onc entra te,
hand le frustration, includ ing ite ms on inatte ntion, hyper activity, a nd impulsivity.
13
A key distinction in delineating treatment effects is to identify symptomatic im-
provement that often precedes behavioral improvement (Burlingame et al., 1995a).
Functional improvements for an adolescent, such as improvements in school
performance and family relationships, often occur later in treatment than do
symptomatic improvements. Because the content areas contained in the Y-OQ
assess various elements of therapeutic change in response to therapy, it is thought
to be sensitive to symptomatic and functional improvements the client is making.
While the Y-OQ stands up well to these criteria, one must always be aware that
questionnaires do not directly measure behavior, situations and moods; they mea-
sure the reporters’ perceptions of those attributes based on responses to questions.
Clinically Meaningful Change
The validity of the Y-OQ rests upon it’s ability to detect change from the previous
assessment. This is especially critical because other popular child assessment
measures such as the Conners’ Parent Rating Scale, the Revised Behavior Prob-
lem Checklist, and the Child Behavior Checklist have not proven adequately
sensitive to measuring changes (Mosier, 1998). Using guidelines established by
Jacobsen and Truax (1991), Y-OQ score intervals have been developed that
indicates normative functioning by the adolescent. When certain cutoff scores are
reached, the client is said to have clinically improved or reached a normal distribu-
tion of symptoms (Wells et al., 1996).
Burlingame et al. (1996) evaluated other inpatient, residential, and outpatient
therapies, and have suggested criteria for assessing whether a client can be labeled
“recovered” or “improved.” Adolescents who have follow-up Y-OQ scores of 46
indicate normal functioning. Therefore, a score of 46 or below is considered
normal, and the client is labeled “recovered.” A client that has moved thirteen or
more points but does not reach the range of normal functioning indicated by a
score of 46, can be labeled “improved.” These two criteria are used to relate the
change in parent assessment and adolescent self-report scores from admission to
discharge.
For example, at admission into treatment, the parent and client are each admin-
istered their respective Y-OQs. The scores generated from these assessments
serve as pretreatment baseline measures of the child’s symptom’s or condi-
tions in the attributes addressed. At the conclusion of treatment, the parent
and client complete questionnaires which assess the same attributes, generating
“discharge” posttreatment scores. The change in scores quantify the client’s
therapeutic progress. For example, a client may enter treatment with a Y-OQ
and SR Y-OQ score of 95 and 100 respectively, and at the termination of the
OBH program, his/her Y-OQ and SR Y-OQ scores may have dropped to 47
and 53 respectively. Relating these scores to the above discussion of clinical
improvement and recovery, a 47 and 53 would be considered clinically “im-
proved” ( a movement of 13 or more points) but not “recovered” because the
score is not 46 or lower.
The Y-OQ has been
shown to be sensitive
to symptomatic and
functional improve-
ments the client is
making.
14
Data Collection and Entry
The parents or legal guardians of clients enrolling at each participating program
between May 1 and December 1, 2000 were asked to sign a research consent
form during the admission process. Of the 1,035 participants in the programs,
858 agreed to participate in the study (83%). Consent forms were administered
and maintained at each respective program. Care was taken by the Y-OQ
administrators at each program to explain the importance of the research in
helping improve OBH treatment. The confidentiality of parents, legal guardians
and clients was maintained through the assignment of a code by each program
administrator, which was used throughout the data collection, analysis and
reporting process.
Clients participating in the study and their parent/legal guardian were asked to
complete the Y-OQ admission questionnaire at admission. For divorced or
separated parents, the primary care parent or legal guardian was asked to
complete the questionnaire. For families with both parents residing in the
household, either parent was asked to complete the questionnaire. If parents or
legal guardians were unavailable, the program administrator mailed them the
questionnaire, along with a return envelope addressed to the University of
Idaho-Wilderness Research Center (UI-WRC) and cover letter outlining the
purposes of the study. An initial phone call was made by each program to
ensure that the parent or guardian received the information, and to answer any
questions regarding their participation in the study.
After each adolescent completed OBH treatment, the parent/legal guardian
was asked to complete a Y-OQ discharge questionnaire. Assessment at
discharge was based on parents’ communication and contact with their child
while he/she was in treatment, communication with the therapist responsible for
his/her care, and contact with their child at graduation ceremonies. The com-
pleted discharge questionnaire was then mailed by the respondent directly to
the UI-WRC.
Clients were also asked to complete a self-report Y-OQ at discharge. This
was done before the client was released from the program to ensure a higher
response rate. These questionnaires were then collected by program adminis-
trators at each site and mailed directly to the UI-WRC. Upon receipt of the
completed Y-OQ questionnaires from each program, a database was con-
structed and a coded file established at the UI-WRC for each client. Ques-
tionnaires were filed according to client identification numbers and program
codes and were accessible only by the principal investigator and a research
assistant.
After each client had completed treatment, the study administrator at each
program mailed the follow-up Y-OQ and SR Y-OQ to each participating client
and their parents or legal guardians at 3-, 6-, and 12-month intervals. A cover
letter accompanied the questionnaire reminding study participants of the impor-
tance of the research and their continued support of the effort. If the client
entered an aftercare program such as an emotional growth boarding school or
residential drug and alcohol treatment center, permission was granted to contact
15
primary care provider at that institution who was asked to encourage the client
to complete the Y-OQ. Aftercare environments included any of the following
types: 1) return to family or primary care giver, 2) transition home, 3) therapeu-
tic boarding school, 4) long term residential drug and alcohol treatment center,
and 5) inpatient hospital.
A UI-WRC addressed envelope accompanied the questionnaire, and the re-
spondents were asked to mail the questionnaire to the UI-WRC when com-
pleted. It is well documented that the percentage of respondents drops consid-
erably at the follow-up time periods (Stinchfield et al.,1994). Because of this,
follow-up phone calls were required to remind participants to complete their
questionnaires.
Study Retention
Retention of participants at follow-up periods was a limitation in this study.
Study retention was not influenced by treatment completion, as 840 clients of
the total of 858 completed treatment, a 97% completion rate. Of the 858
clients who originally agreed to participate in the study, complete assessments
containing an admission and discharge questionnaire were available for 481
(56%) client self-reports and 338 (40%) parent assessments.
Of the 858 clients and parents who agreed to participate in the study, there is at
least one assessment out of a total of five for 737 clients and 594 parents (at
admission, discharge, 3-, 6-, and 12-months). There are several reasons study
participants did not complete questionnaires at various assessment periods.
First, many clients and their parents agreed to participate in the study and
simply did not return questionnaires for reasons unknown. Second, necessary
mail-back procedures were a disadvantage, but were necessary because study
participants came from all regions of the United States. This is despite the fact
that phone calls and follow-up letters were also mailed. Third, clients in after-
care facilities were contacted and sent questionnaires, but did not return them
for unknown reasons. Finally, some clients and their parents had relocated and
could not be found.
Of the 1,035 clients
and their parents
asked to participate
in the study, 858
agreed (83%).
16
12-Month Follow-up Assessment Procedures
Past research on treatment outcomes has shown that non-respondent study partici-
pants at follow-up time periods may have poorer outcome than respondents
(Stinchfield 1994). Researchers have concluded that adolescent outcome reports
that contain a significant number of non-contacted cases may represent overesti-
mates of outcome if generalized to the entire study population. In this study, it was
noted that the number of questionnaires returned to the UI-WRC at the 3- and 6-
month follow-up periods indicated a low response rate, leading to the potential bias
reported above. Therefore, it was determined that the most accurate way to
ensure a more representative sample at the 12-month time period was to contact a
random sample of clients that had at least completed one assessment at admission
or discharge. In doing so, the non-contact bias could be reduced and a more
accurate assessment of client well-being at 12 months after OBH treatment could
be determined.
To accomplish this, a random sample of parents that had completed at least one
assessment at admission or discharge were called at the 12-month time period and
asked to complete a Y-OQ questionnaire over the phone, or by mail. Parents were
selected for contacting because they would also serve as a resource to locate the
adolescent if he or she was not living at home or in an aftercare facility. A total of
594 parents of the original 858 who agreed to participate in the study completed at
least one questionnaire at admission or discharge (70%). To accurately test for
differences in mean treatment scores at discharge and 12-months using a pair-
wise t-test, sample sizes of 138 parents and 78 clients were calculated using a
power equation that would yield a power of .80 at a α=.05 significance level
(Cohen, 1988). The sample size was computed using the standard deviation of the
mean scores at discharge, which was then divided by 13 points, which is the
difference in scores that reflects clinical improvement, yielding an effect size of
.36 for parents and .42 for clients. The list of client coded numbers were randomly
ordered and the first 300 names were selected for contact. (More names were
selected for contact because it was assumed some would be difficult to locate).
Parents who completed a 12-month assessment but did not have completed Y-OQs
at admission and discharge served as a “check” against those that had completed
both admission and discharge questionnaires.
17
A total of 29 parents could not be reached. After phone calls were conducted, a
total of 271 parents completed Y-OQs. Additionally, 139 adolescent clients also
completed Y-OQs after attempted contact based on parent referrals. Table 2
summarizes the results from sampling, contacting, and completed Y-OQs at the 12-
month follow-up period for parents and clients.
Table 2. Total study participants, number who completed at least one Y-OQ during study
period, sample size required to estimate the average at 12-months, total number of study
participants contacted and total number of 12-month follow-up contact calls.
A random sample of
parents that had com-
pleted at least one
assessment throughout
the study period were
called at the 12-
month time period
and were asked to
complete a Y-OQ
questionnaire over the
phone, or by mail.
Total Study
Particpants
Number With
At Least One
Y-OQ
Admission
and
Discharge
Completed
Sample Size
Required Number
Contacted
Completed
Y-OQs at 12-
Month
Parent Y-OQs 858 594 338 138 300 271
Client Y-OQs 858 737 489 78 178 139
18
Data Analysis
Data were analyzed using the Statistical Package for the Social Sciences
(SPSS™). There were several ways that the data could be analyzed. The first
step was to test for normality and to compute averages for scores at each assess-
ment period for both parent and client scores. Confidence intervals (α = .05) were
computed to better understand the range of scores and to determine if any outliers
were present. These data are referred to as the incomplete data set (see Table 3).
Table 3 also shows data from the admission and discharge assessments that were
reported in Technical Report 27, An Assessment of Treatment Outcomes in
Outdoor Behavioral Healthcare (Russell, 2001).
Because client study participation dropped at follow-up periods, data sets were
calculated for each follow-up time period that also included an admission and
discharge score. These data sets are referred to as complete data sets for each
time period and evaluated individual clients at admission, discharge and the respec-
tive follow-up period. For example, a data set was constructed that included those
study participants for whom data were available at admission, discharge and 3-
months. This data set is referred to as the 3-month data set. The first step was to
conduct tests of normality for these data sets at 3-, 6-, and 12-month time periods.
Each data set was found to be normal. Averages and confidence intervals were
also computed for each data set. Paired t-tests were run between discharge
scores and the follow-up time period to examine differences in mean scores.
Finally, data were analyzed to look for differences in average scores at these
assessment points based on client attributes, including gender, age, and aftercare
services utilized.
Non-Response Analysis
To examine the non-response bias at all assessment periods, a series of t-tests
was run to examine differences between complete data sets and all questionnaires
received at that particular time period (incomplete sets) (see Table 3 for results).
This procedure was not necessary for the 12-month data set because a random
sample of study participants were selected. Where statistical differences were
found, it was assumed that the data set was not representative of the full data set
that was received, and could thus be considered biased in the direction of the
identified difference.
19
Two assessments were found to be statistically different: 1) client scores at
admission for the 6-month data set were lower than those for the incomplete
set. This finding suggests that this set of clients may have exhibited fewer
symptoms at admission, and thus would be predisposed to lower scores at
discharge and follow-up periods; and 2) parent scores at discharge for the 6-
month data set were also lower than the incomplete set, suggesting that parents
who continued to complete questionnaires for these clients may have indicated
lower scores at the 6-month follow-up period. Each of these nonresponsive
bias analyses at 6-months is consistent with the literature that suggests that
those study participants that provide data at follow-up periods may exhibit
better outcomes than those clients that do not provide data (see Stinchfield et
al. 1994). Three-month data sets were not found to be statistically different
than the incomplete set, and may be more representative of the sample of
clients. This may be due to the shorter duration of the follow-up contact
period. These analyses and low response rates necessitated the need for the
random sample of clients at the 12-month period.
Table 3. Parent and client Y-OQ scores at admission, discharge, 3-, 6-, and 12-months
arranged by total assessment for that time period, and in 3-month, 6-month and 12-month
data sets.
* Indicates a random sample of study participants was conducted for follow-up period.
1. Averages for this time period were found to be statistically different from the incomplete set
t(719) = -2.579, p <. 01.
2. Averages for this time period were found to be statistically different from the incomplete set
t(409) = -3.371, p <.001.
Admission
(n) Ave. Discharge
(n) Ave. 3-Month
(n) Ave. 6-Month
(n) Ave. 12-M onth*
(n) Ave.
Parent
(Incomplete set) (n=704) 98.72 (n=410) 51.532(n=244) 52.68 (n=247) 50.87 (n=221) 46.64*
Client
(Incomplete set) (n=720) 71.321(n=591) 50.02 (n=155) 47.35 (n=147) 40.48 (n=139) 38.35*
Parent
(Complete discharge set) (n=338) 100.19 (n=338) 48.55
Client
(Complete discharge set) (n=481) 70.67 (n=481) 47.55
Parent
(Complete 3-Month Set) (n=158) 98.99 (n=158) 47.36 (n=158) 51.01
Client
(Complete 3-Month Set) (n=114) 68.69 (n=114) 51.51 (n=114) 46.20
Parent
(Complete 6-Month Set) (n=135) 97.99 (n=135) 42.352(n=135) 50.20
Client
(Full 6-Month Set) (n=103) 63.481(n=103) 46.68 (n=103) 41.07
Parent
(Complete 12-Month Set) (n=144) 97.46 (n=144) 44.94 (n=144) 48.67*
Client
(Complete 12-Month Set) (n=99) 68.30 (n=99) 47.25 (n=99) 38.61*
20
Limitations to the Study
There are other potential sources of error or bias in this study not addressed in
the previous section on non-response bias data analysis. The first limitation to
note is that no control group was utilized and there was no random assignment
of treatment. This is due to the ethical dilemma and cost of establishing control
groups in private placement programs. The mean participation rate among
clients in all programs was 80% (858 of the 1053 clients entering treatment).
Participation rates among programs ranged from 61% to 90% (see Table 4).
Table 4. Total number of clients participating in OBH treatment and participating in the
study from May 1, 2000 to December 1, 2000 in the seven participating programs.
Program Total N umber of Clients
in OBH Treatment Total Number of
Clie nts in Study Percent of
Clie nts in Study
Ascent 217 185 85%
Freer 178 150 84%
SUWS 149 133 90%
Redcliff 140 120 86%
Aspen 172 105 61%
Anasazi 109 86 79%
Sunhawk 70 52 75%
Total Number
of Clients 1035 858 80% Ave.
21
Interpreting Follow-up Results
It is important to note that several factors affect clients once they are dis-
charged from an OBH program. These factors will obviously affect their well-
being, as they try to integrate skills and lessons learned in OBH to posttreat-
ment environments, i.e. aftercare, home, school, and/or peer. It is difficult to
determine the specific factors that are affecting their well-being in these
posttreatment environments. With this said, future research could help to
identify which factors play primary roles in helping clients maintain identified
therapeutic progress. It was simply beyond the scope of this study to do so.
Summary of Limitations
The study findings are based on follow-up data sets from parents and clients,
which represent a percentage of all OBH clients entering treatment in the
seven OBH programs from May 1, 2000 to December 1, 2000. While this
study results represent only those clients agreeing to participate and providing
data sets at respective time periods, it appears as though only the data sets at
6-months may be biased toward clients who exhibited fewer symptoms upon
admission, or were assessed as doing well at discharge by their parents. The
data at discharge, 3-month, and 12-months appears to accurately reflect client
well-being after completion of OBH treatment at these periods.
22
RESULTS
Client Characteristics
Gender and Age
Approximately 70% were of clients were male, and 30% female. Clients
ranged in age from 11-19 years, with 75% of clients between the ages of 16-18
years. Tables 5 and 6 show the study participants’ gender and age.
Table 5. Gender of study participants.
Table 6. Ages of study participants.
The majority of
clients in the
study were males
between the ages
of 16 and 17.
Gender Number of Clients Percent
Male 589 68.6
Female 269 31.4
Total 858 100.0
Age Total Number of Clients Percent
11 1 .1
12 2 .2
13 20 2.3
14 79 9.2
15 159 18.5
16 256 29.8
17 227 26.5
18 81 9.4
19 5 .6
Total 830 96.7
Age Data Not Available 28 3.3
Total Number of Clients 858 100.0
23
Primary Diagnoses
Table 7 shows the types of disorders with which clients were primarily diag-
nosed (according to the DSM-IV manual) and their frequency (some clients
may have been diagnosed with more than one disorder; primary diagnoses only
are reported here). Specific diagnoses were made for 481 of the 858 study
participants (56%) (see Table 7). It is important to note, that due to limitations
in the study, it was not possible to distinguish which clients out of the 377 for
whom no diagnoses were reported did not warrant a diagnosis. It is possible
that a proportion of these clients simply did not warrant a primary diagnosis
after initial assessments by staff at each program. Almost 10% of the diag-
noses were too varied to report here.
Oppositional Defiant Disorder (29%) was the most frequent diagnosis reported,
followed by 25.8% with diagnoses associated with some kind of substance
abuse or dependence (cannabis dependence (10%), cannabis abuse (5%),
alcohol dependence (.7%), alcohol abuse (2.3%), and amphetamine dependence
(1.1%)). Depression Disorder (10%) and Dysthymic Disorder (5%) (a form of
depression), and Bi-Polar Disorder accounted for 22.4% of the diagnoses.
Other primary diagnosed disorders of noted frequency included Adjustment
Disorders (4%) and Bipolar Disorder (3%).
Table 7. Average percentage of primary diagnoses of study participants.
1. Includes Oppositional Defiant, Attention Deficit, and Conduct Disorders.
2. Includes disorders associated with substance abuse or dependence.
3. Includes Depression, Dysthymia and Bipolar Disorders.
4. Includes Anxiety and Adjustment Disorders.
Disorder Number of Clients Percent
Behavio ral D isorders1182 37.8
Substance Disorders2145 30.1
M ood Disorders3108 22.4
Other
445 9.7
Total 481 100
Diagnoses Available 481 56.1
N o Diagnoses Data Available 377 43.9
24
Prior Treatment
Previous outpatient services were received by more than half of all OBH
clients (57%). Study participants had also previously received inpatient treat-
ment services before enrolling in OBH (17.4%). Thirteen percent of all clients
received both inpatient and outpatient services; this group represents perhaps
the most seriously affected clients in the study. Inpatient services consists of
those services where the client was in protective care of therapeutic facility,
while outpatient services, which consisted of periodic individual or group
counseling sessions, are those services where the child remained in the protec-
tive custody of the parent.
Table 8. Frequency and percentage of clients who received inpatient and outpatient
treatment services prior to enrolling in OBH treatment.
Summary of OBH Client Characteristics
Clients who participated in the study demonstrated the following characteris-
tics: 1) the majority of the 858 clients were male (69%) and were between the
ages of 16-18 years old (75%) (see Tables 5 and 6); 2) client diagnoses in-
cluded Behavioral Disorders (38%), substance disorders (30%), and depressive
disorders (22%) (Table 7); and 3) over half of the OBH clients (57%) had
received outpatient services prior to enrolling in an OBH program, 17% had
received inpatient treatment, and 13% had utilized both outpatient and inpatient
prior to OBH treatment (Table 8).
Table 8 shows that
57% of clients had
received outpatient
counseling prior to
OBH treatment.
Prior Inpatient Services Number of Clients Percent
Yes 149 17.4
No 709 82.6
Total 858 100
Prior O utpatie nt Se rvice s
Yes 491 57.2
No 367 42.8
Total 858 100
Prior Inpatient and Outpatient Services
Yes 115 13.4
No 743 86.6
Total 858 100
25
TREATMENT OUTCOME
Research Question 1. What OBH treatment outcomes were still
present at 3-, 6-, and 12-month follow-up periods?
Overall, the results indicated that the OBH clients participating in the study had
reduced behavioral symptoms at discharge as measured by both client self-
reported and parent assessments (Table 8). For client self-reports, group
means decreased 23.12 points from 70.67 to 47.55 between admission and
discharge. Parent assessments decreased 51.64 points from 100.19 at admis-
sion to 48.55 at discharge, more than twice the improvement reported by
clients. These results suggest that the average OBH adolescent client reported
clinically significant improvement after OBH treatment, and had outcome
scores at discharge that were within two points of a normal range of function-
ing (score of 46 or below), similar to a normed sample of adolescents their age
(Burlingame et al., 1996).
Table 8. Admission and discharge average SR Y-OQ and Y-OQ scores, including mean
differences in scores from admission to discharge. Average differences in
client self-report scores
from admission to dis-
charge showed a more
than 20 point difference,
suggesting clinically
significant improvement
from treatment.
NPeriod
Average M ean
Score
(sd*)
Mean
Difference
Client Self R eport
SR Y-OQ 481 Admission 70.67
(32.86) 23.12
481 Discharge 47.55
(31.23)
Parent Assessment
Y-OQ 338 Admission 100.19
(28.52) 51.64
338 Discharge 48.55
(37.48)
* Indicates standard deviation of scores. Other assessments using the Y-OQ
report similar standard deviations. For example, Mosier et al. (2001) report
standard deviations from 36.68 to 40.521 at various assessment periods.
26
Outcomes at 3-Month Follow-up Period
Outcome data available at the 3-month follow-up period indicate that clients
maintained progress resulting from OBH treatment. Client self-report scores
show a discharge score of 51.51 and a 3-month score of 46.20 (n=114) (see
Table 10). The incomplete client self-report data set shows a discharge score
of 49.10 and a slightly higher score at 3-months of 50.54, suggesting that for
this smaller sample of clients, outcomes were also maintained. Parent assess-
ments show a slightly higher score at 3-months of 51.01, increasing 4-points
from 47.36 at discharge (n=158). For both client self-reports and parent
assessments, scores were higher at the 3-month follow-up period for the
incomplete data set, suggesting that clients for whom a complete data set was
available at admission, discharge and 3-months had better outcomes than those
that did not report scores. Paired t-tests showed no significant differences
between scores at discharge and the three-months for both the incomplete and
the 3-month data sets.
The 95% Confidence Interval for c client scores for the complete 3-month data
set ranged from 38.62 to 50.66 for client self-reports and 39.53 to 53.20 for
parent assessments (α = .05). These confidence intervals suggest that clients
maintained outcomes from treatment at the 3-month follow-up period, with 95% of
the scores falling within 4-7 points of the normal cut-score score of 46 estab-
lished by Burlingame et al. (1996).
Table 10. Average SR Y-OQ and Y-OQ scores and frequency for the incomplete data set
and the 3-month data set.
Average Admission
(n) Ave.
(sd)
Average Discharge
(n) Ave.
(sd)
Average Three-Month
(n) Ave.
(sd)
Client Self R e port
Incomplete Set (n= 606) 71.87
(31.85) (n=477) 49.10
(32.56) (n=41) 50.54
(30.67)
Three-Month Set (n= 114) 68.69
(32.72) (n= 114) 51.51
(33.12) (n= 114) 46.20
(31.05)
Parent Assessment
Incomplete Set (n= 546) 99.04
(27.61) (n= 252) 52.80
(35.21) (n= 86) 55.70
(36.47)
Three-Month Set (n= 158) 98.99
(28.81) (n= 158) 47.36
(36.92) (n= 158) 51.01
(38.25)
27
Outcomes at 6-Month Follow-up Period
At 6-months, clients self-report scores had dropped from 46.68 at discharge to
41.07 at 6-months (n=103 clients). This finding indicates that these clients
viewed themselves as continuing to make progress after termination of the
OBH treatment programs. Parents assessed clients at this period with a score
of 42.35 at discharge and 50.20 at 6-months (n=135). Though this suggests an
deterioration of outcomes, scores were not significantly different than the cut-
score of 46. The differences in parent scores were not statistically significant
(p. > .20).
The 95% Confidence Interval for client scores for the complete 6-month data
set ranged from 34.77 to 47.37 for client self-reports and 44.24 to 56.16 for
parent assessments (α = .05). These confidence intervals also suggest that clients
maintained outcomes from treatment at the 6-month follow-up period, with 95% of
the scores falling within 4-10 points of the normal cut-score score of 46 estab-
lished by Burlingame et al. (1996).
Differences between the 6-month data set for clients and parents were found
when compared to the incomplete data set, suggesting that clients for whom
complete data were available may have had fewer symptoms than those clients
in the incomplete set. Parent assessments for the 6-month data set show dis-
charge scores that were statistically different from the incomplete set (t(409) =
-3.371, p <. 001); this suggests that these clients may have responded more
favorably to treatment.
Table 11. Average SR Y-OQ and Y-OQ scores and frequency for the incomplete data set
and the 6-month data set.
Average Admission
(n) Ave.
(sd)
Average Discharge
(n) Ave.
(sd)
Average Six-Month
(n) Ave.
(sd)
Clie nt Se lf Report
Incomplete Set (n= 617) 72.89
(25.10) (n=477) 49.10
(32.52) (n=44) 39.09
(31.27)
Six-Month Set (n= 103) 63.48
(32.07) (n= 103) 46.68
(35.83) (n= 103) 41.07
(32.24)
Parent Assessment
Incomplete Set (n= 569) 99.04
(26.12) (n= 275) 52.80
(28.37) (n= 112) 56.69
(34.26)
Six-Month Set (n= 135) 97.99
(27.10) (n= 135) 42.35
(35.64) (n= 135) 50.20
(34.99)
28
Figure 2 shows a box plot illustrating the range of responses for all clients at
admission, discharge, and at 3- and 6-month follow-up periods, and offers insight
into client well-being after treatment. Figure 2 shows median scores for client
self-report scores below the cut score of 46 for 3- and 6-month assessment
periods. At three-months posttreatment, approximately 25% of all clients had
scores between 25-45, while the other 25% had scores that ranged from 45-70. At
six-months posttreatment, the range of the “box” tightened, with 50% of the scores
falling between 25 and 55 (this means that 50% of scores at this time period would
not statistically differ from the cut score of 46) .
Figure 2. Boxplot depicting the range of client self-report Y-OQ scores at admission,
discharge, 3-, and 6-months (Line denotes normal score of 46).
Figure 3 shows parent scores at admission, discharge, and 3- and 6-month
follow-up periods. Parents’ scores indicate a median at 3- and 6- month
follow-up periods that is also close to the cut score of 46. The “boxes” remain
consistent in their distribution of scores at 3- and 6-months, with scores ranging
from 20-70.
Figure 3. Boxplot depicting the range of parent assessed Y-OQ scores at admission,
discharge, 3-, and 6-months (Line denotes normal score of 46).
147155591720N =
Asessm ent Period
6-M onth3-M onthDischargeAdm ission
Y-O Q Score
200
150
100
50
0
-5 0
-1 0 0
247244410704N =
Assessm ent Period
6-M onth3-M onthDischargeAdm ission
Y-O Q Score
200
150
100
50
0
-5 0
-100
Box plots: The upper and
lower boundaries of the box
are drawn at the 75th and
25th percentiles. The box
represents 50% of all scores
in the distribution. The line
in the center of the box is the
median, where 50% of the
scores lie above or below the
line. The lines extending
from the box reach from the
extreme highest to the
extreme lowest score and
represent less than 1% of the
scores.
29
Outcomes at 12-Month Follow-up Period
The data at 3- and 6-month assessments show outcomes from treatment had
been maintained by clients; however, these data represent a self-selected
sample that could be biased toward more favorable outcomes. To address this,
a random sample of clients who had completed at least one assessment at
admission or discharge were selected to ensure a more representative sample
at the 12-month time period. In doing so, the non-contact bias was reduced and
a more accurate assessment of client well-being at 12-months after OBH
treatment was determined.
Table 12 shows that a total of 271 parents and 139 clients completed the Y-OQ
questionnaire at the 12-month time period. The next process was to identify study
participants who completed questionnaires at both admission and discharge. This
provided data sets that yielded information regarding treatment progress (differ-
ence in scores from admission to discharge) and the resulting outcome at the 12-
month time period. Clients’ self report outcomes averaged 8 points under the cut-
score of 46 (38.61), indicating that those clients maintained outcomes from treat-
ment, and actually continued to improve at the 12-month follow-up period. Parent
scores were almost 10 points higher (48.67), but were not significantly different
than the cut-score of 46 points. Client self-reports and parent assessments indi-
cated that OBH graduates maintained outcomes from treatment and were doing
well emotionally and behaviorally at the 12-month follow-up. T-tests were run on
the averages from the incomplete and complete data sets at all time periods,
and significant differences were only found for parent assessed discharge
scores (t(409) = -2.519, p < .012). This finding indicates that the randomly
sampled 12-month data set had lower discharge scores than the incomplete
data set (a difference of 10.16), but also had higher scores at the 12-month
follow-up period (a difference of 5.83).
Table 12. Average SR Y-OQ and Y-OQ scores and frequency for the incomplete data set
and the 12-month data set.
Average Admission
(n) Ave.
(sd)
Average Discharge
(n) Ave.
(sd)
Average Twelve-M onth
(n) Ave.
(sd)
Client Self R eport
Incomplete Set (n= 621) 71.80
(28.38) (n=492) 50 .58
(34.72) (n=40) 37 .70
(30.61)
Twelve-M onth Set (n= 99) 68.30
(34.14) (n= 99) 47 .25
(30.78) (n=99) 38.61
(31.83)
Parent Assessment
Incomplete Set (n= 560) 99.04
(27.65) (n= 266) 55.10
(32.28) (n= 127) 42 .84
(30.29)
Twelve-M onth Set (n= 144) 97.46
(28.81) (n= 144) 44.94
(36.92) (n= 144) 48.67
(38.25)
30
A data set was constructed that included both parent and client assessments
for an individual client to further explore the more than 10-point difference that
was found between client self-reports and parent assessments at the 12-month
time period. Figure 4 shows box plots for client (c) and parent (p) scores at
admission, discharge and the 12-month follow-up period for 61 randomly
sampled clients for whom assessments were made by both parents and clients.
Though average admission scores were noticeably higher for parents than
clients (99.36 and 69.82 respectively) (see Russell, 2001, for suggestions as
why they may have been differences in scores at admission), scores at dis-
charge for parents (46.59) and clients (43.77) were similar. At the 12-month
follow-up period, scores were also similar, with parent assessments averaging
43.70 and clients averaging 43.34. This suggests that parents and clients were
rating the adolescent consistently using the Y-OQ at follow-up periods, and that
outcomes were maintained for this randomly selected data set 12-months after
completion of treatment.
Figure 4. Boxplot depicting the range of parent assessed Y-OQ scores at admission,
discharge, and 12-months (Line denotes normal score of 46).
616161616161N =
A ssessm ent Period
(P ) 12-M onth
(P ) Discharge
(P ) Adm ission
(C ) 1 2 -M o n th
(C ) D ischarge
(C ) A d m ission
Y-OQ Score
200
150
100
50
0
-5 0
-100
31
Outcomes at 12-Month Follow-up Period According to
Gender and Age
Outcomes at the 12-month follow-up period were further analyzed to explore
differences in total Y-OQ score by age and gender. Table 13 shows that
females entered treatment with more severe presenting symptoms as indicated
by both client self-report and parent assessment. Females self-reported an
admission score that was 26 points higher than males, and a discharge score
that was also higher by more than 13 points. Males self-reported discharge
scores below the cut-score of 46 (42.0), while females remained significantly
higher than the discharge score (55.68). At the 12-month follow-up period,
females continued to show improvement indicated by a score of 40.42, which
was below the cut-score of 46. Males also improved by more than four points
to a score of 37.48. At 12-months, both males and female self-report scores
were below the cut-score of 46. Females continued to show improvement
between discharge and 12-months evidenced by a significantly different drop in
scores from 55.68 at discharge to 40.42 at 12-months (t(41) = 2.384, p<.02).
Parent assessments show similar patterns in the maintenance of outcomes
throughout the follow-up period, though scores did rise for males and females
between discharge and 12-months by 3 and 4 points respectively. Scores at 12-
months remained 3 points above the cut score of 46 for males (49.66) and at
the cut-score for females (46.54).
Table 13. Average SR Y-OQ and Y-OQ scores and frequency for the 12-month data set for
male and female study participants.*
Ave. Admission
(n) Av e. Ave. Discharge
(n) Av e. Ave. 12-M onth
(n) Av e.
Client Self Report
Male (n = 61 ) 58.05 (n = 61 ) 42.00 (n = 61 ) 37.48
Female (n = 38 ) 84.76 (n = 3 8) 55.68 (n = 3 8) 40.42
Parent Assessment
Male (n = 98 ) 94.05 (n = 98 ) 46.32 (n = 98 ) 49.66
Female (n = 46 ) 104.72 (n = 46 ) 42.02 (n = 46 ) 46.54
*Standard deviations ranged from 22.75 to 38.67 for all scores in Table 13.
32
Table 14 shows clients ages 13-14 (n=12) reported the highest admission
scores across all age groups. Discharge scores for this age group also re-
mained high and were more than 17 points above the cut-score of 46. Clients
ages 17-19 (n=34) reported the lowest discharge scores by more than 14
points when compared to clients ages 15-16 (n=49); these scores were 10
points below the cut-score of 46. Client self reports for all age groups were
below the cut score of 46, indicating that outcomes at discharge were main-
tained. Clients ages 13-14 continued to improve during this time period, indi-
cated by a drop in scores of more than 25 points from discharge to the 12-
month follow-up period.
Parent assessments showed higher scores at admission for clients ages 13-14
(n=16) when compared to other age groups. At discharge, parents assessed
all clients at or below the cut-score of 46, with clients ages 17-19 (n=52)
showing the most improvement in treatment and lowest discharge score
(39.31). Parents indicated a 3-5 point increase in scores from discharge at 12-
months, with clients ages 15-16 showing an increase in scores to above the cut-
score of 46.
Table 14. Average SR Y-OQ and Y-OQ scores and frequency for the 12-month data set for
clients classified into three age groups: ages 13-14, ages 15-16, ages 17-19.*
Ave. Admission
(n) Av e. Ave. D is charge
(n) Av e. Ave. 12-Month
(n) Av e.
Client Self-Report
Ages 13-14 (n=12) 72.50 (n=12) 63.50 (n=12) 37.75
Ages 15-16 (n=49) 66.88 (n=49) 50.86 (n=49) 42.88
Ages 17-19 (n=34) 69.24 (n=34) 36.21 (n=34) 34.24
Parent Assessment
Ages 13-14 (n=16) 107.19 (n=16) 45.31 (n= 16) 48.31
Ages 15-16 (n=71) 96.45 (n=71) 46.37 (n=71) 51.15
Ages 17-19 (n=52) 95.10 (n=52) 39.31 (n=52) 44.71
*Standard deviations ranged from 24.26 to 39.14 for all scores in Table 14.
33
Outcomes at 12-Month Follow-up Period According to
Utilization of Aftercare Services
Consideration of the utilization of aftercare services by clients is important in
understanding OBH treatment outcomes. Because OBH programs are often
short term interventions (average length of treatment in this study was 45
days), it is often recommended that clients enroll in an aftercare program to
help them maintain therapeutic progress. Clients were categorized into two
groups: 1) utilizing aftercare services, which were defined as including place-
ment of the child in a residential facility outside of the home environment for
between three and 12 months, and 2) returning to the home environment. It was
beyond the scope of this study to further break down the aftercare service
classifications.
Table 15 shows that 45.3% of the clients who participated in the study utilized
aftercare services, while over 50% of all clients returned home after comple-
tion of treatment. Data was not available for 4.2% of the study participants.
Table 15. Number and percentage of study participants that participated in and did not
participate in aftercare services.
A fte rc are
Serives Frequency Percent
Yes 389 45 .3
N o (H ome) 433 50.5
Not
Available 36 4.2
Total 858 10 0
34
Table 16 reports data from all clients (incomplete data set) and shows no
significant score differences at the 12-month follow-up period for those clients
who utilized aftercare services and those that returned home for both client
self-reports and parent assessments. A paired t-test was run to examine the
more than 9 point difference in parent assessments at this period (52.83 and
43.17). No significant differences were found t(196) = -1.769, p >.08).
Table 16. Total Y-OQ scores at admission, discharge and 12-months for clients who
utilized aftercare services and those who returned home.*
*Standard deviations ranged from 22.42 to 40.61 for all scores in Table 16.
Table 17 reports complete data sets and shows no statistical differences in
scores between those clients who utilized aftercare services and those who
returned home. There were however, real differences in scores for both client
self-report and parent assessments. Client self-report data show that clients
who utilized aftercare services had higher scores at admission than those that
returned home by more than 8 points (72.63 and 63.98) and higher discharge
scores by almost 5 points (51.80 and 46.89). At the 12-month follow-up period,
clients who utilized aftercare services had higher scores by more than 6 points
(43.39 and 37.17). This data suggests that clients who utilized aftercare
services remained above the normal cut score of 46 at discharge and programs
and parents may have believed they were still at-risk of resorting to past
behaviors. Both groups indicated self-report scores below the cut score of 46.
Admission
(n) Ave. Discharge
(n) Ave. 12-Month
(n) Ave.
AFTERCARE Client Self-Report Y-OQ (n=307) 73.03 (n=263) 52.42 (n=52) 41.29
Parent Assessment Y-OQ (n=301) 100.43 (n=158) 50.63 (n=99) 43.17
HOME Client Self-Report Y-OQ (n=350) 69.59 (n=282) 48.84 (n=70) 37.90
Parent Assessment Y-OQ (n=331) 98.34 (n=207) 53.32 (n=99) 52.83
35
*Standard deviations ranged from 27.29 to 37.11 for all scores in Table 17.
Parent assessments show a different pattern of scores for these two groups.
For the aftercare group, scores were higher at admission (100.18 and 95.71)
and slightly lower at discharge (44.79 and 46.93). At the 12-month follow-up
period, scores were also lower for the aftercare group (47.62 and 52.13).
Table 17. Average Y-OQ scores for complete data sets at admission, discharge, and 12-
months for clients who utilized aftercare services and clients who returned home.
Admission
(n) Ave. Discharge
(n) Ave. 12-Month
(n) Ave.
AFTERCARE Client Self-Report Y-OQ (n=41) 72.63 (n=41) 51.80 (n=41) 43.39
Parent Ass essment Y-OQ (n=56) 100.18 (n=56) 44.79 (n=56) 47.62
HOME Client Self-Report Y-OQ (n=47) 63.98 (n=47) 46.89 (n=47) 37.17
Parent Ass essment Y-OQ (n=70) 95.71 (n=70) 46.93 (n=70) 52.13
36
Outcomes at 12-Month Follow-up Period: Differences in Six Content
Areas
The six content areas of the Y-OQ are designed to assess symptoms associ-
ated with: (1) Interpersonal Distress, (2) Somatic, (3) Interpersonal Relations,
(4) Critical Items, (5) Social Problems, and (6) Behavioral Dysfunction. These
content areas are referred to as subscales in the Y-OQ and are outlined in
detail in Figure 1. There are two possible ways in which subscale data can be
analyzed: 1) conduct pair-wise comparisons between discharge and 12-months
to look for significant differences in scores, and 2) examine the “cut scores”
associated with each subscale to determine if 12-month scores were at or
below these cut scores (Burlingame et al., 1996). This would indicate a return
to a normal range of symptoms in each content area.
Three subscales were found to be significantly different between discharge
and 12-months for client self-reports. They were: (2) Somatic (improvement
made in scores), (3) Interpersonal Relations (deterioration in scores), and (6)
Behavioral Dysfunction (improvement made in scores). One subscale was
found to be significantly different between discharge and 12-months for parent
assessments: (3) Interpersonal Relations. Scores for subscales (3) Interper-
sonal Relations and (6) Behavioral Dysfunction were above the cut-score for
both client self-report and parent assessment.
Table 18. Y-OQ average subscale scores at discharge and 12-months, including cut-
scores established by Burlingame et al. (1995) for each of the six content areas.
Frequency Ave rage
Discharge Average
12-Month Cut-Score
Client Self-Report
Subscore 1 (Interpersonal Distress) 99 15.99 16.51 16.4
Subscore 2 (Somatic) 99 5.58 4.55 5.0
Subscore 3 (Interpersonal Relations) 99 3.88 5.52 4.4
Subscore 4 (Critical Items) 99 5.20 5.03 5.0
Subscore 5 (Social Problems) 99 12.02 12.99 12.0
Subscore 6 (Behavioral Dysfunction) 99 6.98 5.35 3.0
Parent Assessment
Subscore 1 (Interpersonal Distress) 144 17.16 17.61 16.4
Subscore 2 (Somatic) 144 3.45 4.32 5.0
Subscore 3 (Interpersonal Relations) 144 5.96 7.77 4.4
Subscore 4 (Critical Items) 144 5.47 4.49 5.0
Subscore 5 (Social Problems) 144 10.25 10.89 12.0
Subscore 6 (Behavioral Dysfunction) 144 5.02 5.89 3.0
37
DISCUSSION
Study participant outcomes at 3-, 6-, and 12-month follow-ups indicate that
clients are maintaining outcomes from OBH treatment at follow-up periods.
The data from the random sample of study participants at the 12-month
follow-up period suggest that clients are maintaining outcomes, indicated by
Y-OQ and SR Y-OQ scores that, on average, continue to decline after
completion of treatment. In general, results suggest that clients are making
the transition from the wilderness environments of OBH into posttreatment
environments successfully, be they home or aftercare programs. Future
research could examine the factors that make this transition easier for the
client and parents to better understand why certain clients may do better in
this transition than others. However, data at the 3- and 6-month follow-up
period may include a self-selected sample of clients who may be doing
better than non-respondents.
Clients for whom data was available at admission, discharge and 12-months
reported average improvement from 47.25 (slightly above the normal score
of 46) to 38.61 at 12-months after completion of treatment. Interestingly,
parents reported scores of 44.94 at discharge and 48.67 at 12-months,
suggesting a slight deterioration of effects. Twelve-month scores for clients
and parents are also statistically different (t(220) = 18.49, p < .001). Parents
and clients differed by more than 30 points in average scores at admission,
yet had similar scores at discharge, and at 3-, and 6-month time periods but
began to differ again at the 12-month follow-up period. Perhaps adolescent
clients believe they are doing well at this point in time, but parents are again
becoming concerned about their child’s behaviors and the feelings their child
is experiencing. The differences between parent and client assessments at
various periods of treatment is also an area for further research.
Due to limitations in this study, specific OBH treatment process factors that
help facilitate this maintenance of outcomes is unknown. It is also beyond
the scope of this study to assess which OBH-acquired skills--be they
behavioral or interpersonal--most contribute to the maintenance of out-
comes. How clients apply skills and lessons learned in OBH and related
programs to their everyday lives has not been well documented in the
literature (Hattie et al., 1997; Winterdyk & Griffiths, 1984). Some studies
suggest certain process factors that may be beneficial in certain posttreat-
ment environments. These include working with the parents and child to help
improve channels of communication (Bandoroff & Scherrer, 1994); having
the adolescent experience something as physically and emotionally demand-
ing as a wilderness expedition, and thus realizing that other challenges in life
may not be as intimidating (Russell, 2001); or experiencing an intense
interpersonal experience through group development and cohesion that is
desirable and something to strive for outside of wilderness experiences
(Ewert & Heywood, 1991; Russell, 2002). Future research could address
what specific process factors, skills and lessons are being developed in
OBH treatment and assess how these relate to outcomes in posttreatment
environments.
38
Gender and age were important factors that explained variance in scores in this
study. Client self-reports indicated an average admission score of 58.05 for
males and 84.76 for females, a difference in score of more than 26 points
(t(97) = 4.077, p < .001). Scores remained different at discharge by more than
13 points (55.68 and 42.00), but were similar at 12-months, with a difference of
less than 3 points (40.42 and 37.48). (Parent assessments showed higher
scores at admission for females and below the cut-score and lower than males
at discharge and 12-months). Burlingame et al. (1996) examined gender
differences in Y-OQ scoring, and found no reliable differences between males
and females in total Y-OQ scores, but did find differences in two subscales.
Male were found to have higher behavioral dysfunctional scale scores than
females, while females have higher somatic scale scores than males (p. 8).
This may explain score differences between male and females. There are two
observations from these findings. First, it is unclear why significant differences
exist in self-report scores for males and females. Do females respond better
to OBH treatment than males? Do females report higher rates of certain
symptoms at admission than males? Second, higher scores at admission for
females indicated by parent assessments may suggest that females may have
more severe presenting symptoms than males. These are interesting findings
that could be the focus of further research.
Client self-reports and parent assessments both indicate that older clients (ages
17-19) seem to have more favorable outcomes from OBH treatment than
younger clients (ages 15-16). However, the youngest clients were shown to
have the most severe presenting issues. Admission scores for both client self-
report and parent assessments were 6 and 11 points higher than other age
groups respectively. The youngest clients, ages 13-14, reported considerable
improvement at the 12-month follow-up period, with a reduction in scores of
more than 25 points during this time. Parent assessments showed a slight
increase in scores for all ages from discharge to 12-months.
No statistical differences were found in average scores when comparing
clients who utilized aftercare services with those who returned home. How-
ever, there were noticeable score differences. Client self-report data show that
clients who utilized aftercare services had higher scores at admission and higher
discharge scores. At the 12-month follow-up period, scores were still higher by
more than 6 points. This may suggest that aftercare services may be warranted
for these clients who had made progress in treatment but still needed the structure
of aftercare services. Parent assessments show a different pattern of scores. For
the aftercare group, scores were higher at admission and slightly lower at dis-
charge and were also lower at the 12-month follow-up period. This is also an
important area for further research. It may be possible to utilize a random sample
of clients who would either go on to an aftercare program or return home. Differ-
ences in outcomes for these two groups could then be compared to better under-
stand the role aftercare services may play in maintaining outcomes.
39
Analysis of subscale data showed a deterioration of outcomes in interper-
sonal relations at the 12-month follow-up period that were above the cut
score, and statistically different than scores at discharge. Client self-report
scores at discharge were below the cut-score, suggesting that clients may
have made progress in this area from group interaction and peer relations,
and through relationships established with therapeutic staff. These are two
powerful factors in all therapeutic modalities that have been suggested to
be enhanced through group living in wilderness environments (Ewert &
Heywood, 1991; Russell, 2001 and 2002). The deterioration of outcomes in
this area may reflect inabilities to have these types of interpersonal con-
nections to peers and adults in posttreatment environments. Also, clients
self-reported a significant improvement in the behavioral dysfunction
content area, which may be an indication of improvement in organizing
tasks, completing assignments in school, and learning how to handle frus-
tration in more appropriate ways.
40
SUMMARY AND CONCLUSIONS
Conclusions from the first phase of this study reported in Technical
Report 27 (Russell, 2001) are:
1. The clients enrolling in the seven OBH programs during the period of this
study were predominantly male (69%) and between the ages of 16-18 (75%).
Ninety-seven percent of clients who entered OBH programs and participated
in the study completed treatment.
2. Clients entered treatment with a variety of disruptive behavioral, mood
and substance disorders as their primary diagnoses. The most frequent
disorders for those with diagnoses (56% of the total study population) were
behavioral disorders (38%), including Oppositional Defiant, Attention Deficit,
and Conduct Disorder. Substance Disorders (includes a wide range of
substances, including alcohol, cannabis, and tobacco) comprised 31% of
diagnoses, while mood disorders comprised 23%, which includes Depression
or Dysthymic diagnoses.
3. Over half (57%) of the OBH clients had received outpatient services,
while 17% had utilized inpatient services prior to OBH treatment. Many
(13%) had tried both types of services prior to OBH treatment.
4. Clients moved toward a normal range of symptoms after completion of
OBH treatment. Client self-report mean Y-OQ scores at admission were
70.67; mean discharge scores were 47.55, indicating an average score reduc-
tion of more than 20 points. Parent assessed Y-OQ mean scores at admis-
sion were 101.19 and their mean discharge scores were 48.55, indicating an
average reduction of 52.64 points.
5. OBH has positive outcomes when compared to two recent studies re-
ported in the literature of in-home, family centered psychiatric treatment
(Mosier et al., 2001) and a partial-day treatment program for referred chil-
dren (Robinson, 2000) that had also measured treatment outcomes using the
Y-OQ.
Conclusions from the second phase of this study assessing outcomes
from OBH treatment at 3-, 6-, and 12-month follow-up periods are:
1. At the 3- and 6-month follow-up period, outcomes were maintained as
indicated by client self-report scores that did not significantly differ from
previous assessments; parent assessments indicated higher scores at 3-month
(4 points) and 6-month (8-points) time periods when compared to discharge
scores, suggesting a deterioration of outcomes. Based on the supporting
literature, differences in scores between complete and incomplete data sets,
and the small sample size of clients for whom data was available at follow-up
periods, these samples may be biased in the direction of more favorable
outcomes. Thus, the decision was made to conduct a random sample of
clients at the 12-month follow-up period in hopes of gaining a more represen-
tative and unbiased sample.
41
2. Client self-report and parent assessment at the 12-month follow-up period
indicate that clients maintained outcomes from treatment and were doing well
emotionally and behaviorally. Clients self reported outcomes that averaged 8
points below the cut-score of 46 (38.61), indicating that those clients had actually
continued to improve up to one-year after completion of treatment. Parent
scores were almost 10 points higher at 48.67, but were also close to the cut-
score of 46 points. These findings suggest that clients had maintained behavioral
and emotional outcomes from treatment at the 12-month follow-up period.
3. For the randomly sampled data set at 12-months (admission, discharge,
and 12-months), both males and females self-reported below the cut-score of
46. Females report a higher level of symptoms at admission, indicated by a
score that was more than 26 points higher than males. Parent assessments
show similar patterns in the maintenance of outcomes throughout the follow-
up period, though scores did rise for males and females between discharge
and 12-months by 3 and 4 points, respectively.
8. For the randomly sampled data set at 12-months (admission, discharge,
and 12-months), clients ages 13-14 self-reported the highest admission scores
across all age groups, and also had higher discharge scores. Older clients
(ages 17-19) self-reported the lowest discharge scores. Younger clients
showed statistically significant improvement between discharge and 12-
months, evidenced by a drop in scores of more than 25 points. Parent as-
sessments showed similar patterns across age groups, but also indicated an
increase in scores from discharge to the 12-month period by five points above
the cut score (46).
9. No statistical differences were found in average scores when comparing
clients who utilized aftercare services with those who returned home. How-
ever, client self-report data show that clients who utilized aftercare services
had higher scores at admission and discharge than those clients who returned
home. At the 12-month follow-up period, clients who utilized aftercare
services had higher scores by more than 6 points. This data suggests that
clients who utilized aftercare services remained above the normal cut score
of 46 at discharge; OBH program staff and parents may have believed they
were still at-risk of resorting to past behaviors. Parent assessments show
that the aftercare group had higher scores at admission but were slightly
lower at discharge and the 12-month follow-up period.
42
LITERATURE CITED
Achenbach, T. M. (1991). Manual for the child behavior checklist 14-18
and 1991 profile. Burlington: University of Vermont, Department of
Psychiatry.
American Psychiatric Association. (1994). Diagnostic manual of mental
disorders. (4th ed.). Washington, DC: Author.
Bandoroff, S. (1989). Wilderness therapy for delinquent and pre-delinquent
youth: A review of the literature. (ERIC ED377428): University of
South Carolina, Columbia, SC.
Bennett, L., Cardone, S., & Jarczyk, K. (1998). Effects of a therapeutic camp-
ing program on addiction recovery: The Algonquin Haymarket relapse
prevention program. Journal Of Substance Abuse Treatment, 15(5),
469-474.
Block, J. H. & Robins, R.W. (1993). A longitudinal study of the consistency
and change in self-esteem from early adolescents to early childhood.
Child Development, 64, 909-923.
Brownell, K. D., Marlatt, G. A., Lichtenstein, E., & Wilson, G. T. (1986).
Understanding and preventing relapse. American Psychologist, 41,
765-782.
Burlingame, G., Lambert, M., Reisenger, C., Neff, W., & Mosier, J. (1995a).
Pragmatics of tracking outcomes in a managed care setting. The Jour-
nal of Mental Health Administration, 22(3), 226-236.
Burlingame, G. M., Wells, M. G., & Lambert, M. J. (1995b). The Youth Out-
come Questionnaire. Stevenson, MD: American Professional
Credentialing Services.
Burlingame, G. M., Wells, M. G., Hoag, M. J., Hope, C. A., Nebeker, R. S.,
Konkel, K., McCollam, P., & Reisenger, C. W. (1996). Manual for
youth outcome questionnaire (Y-OQ). Stevenson, MD: American
Professional Credentialing Services.
Burton, L. M. (1981). A critical analysis and review of the research on Out-
ward Bound and related programs. Dissertation Abstracts Interna-
tional, 47/04B (University Microfilms No. AAC812247).
Cason, D. R., & Gillis, H. L. (1994). A meta analysis of adventure program-
ming with adolescents. Journal of Experiential Education, 17, 40-47.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences.
(2nd ed.). Hillsdale, NJ: Erlbaum.
43
Davis-Berman, J., & Berman, D. S. (1994). Wilderness Therapy: Founda-
tions, theories and research. Dubuque, IA: Kendall/Hunt Publishing.
Deschenes, E. P., & Greenwood, P. (1998). Alternative placements for juvenile
offenders: Results from the evaluation of the Nokimus Challenge
Program. Journal of Research in Crime and Delinquency, 35(3),
267-294.
Easley, A. T., Passineau, J. T., & Driver, B. L. (1990). The use of wilderness
for personal growth, therapy, and education. (General Technical
Report RM-193). Fort Collins, CO: Rocky Mountain Forest and Range
Experiment Station.
Eggleston, E. J. (1998). Reflections on wilderness therapy. In: Exploring the
Boundaries of Adventure Therapy: International Perspectives. Itin, C.
(Ed.). Proceedings of the 1st International Adventure Therapy Confer-
ence: Perth, Australia, July 1997. Association for Experiential Educa-
tion, Boulder, CO.
Ewert, A, & Heywood, J. (1991). Group development in the natural environ-
ment: Expectations, outcomes and techniques. Environment and
Behavior, 23, 5, 592-615.
Ewert, A. (1983). Outdoor adventure and self concept: A research analy-
sis. Eugene: University of Oregon: Center for Leisure Studies.
Ewert, A. (1987). Research in experiential education: An overview. Journal of
Experiential Education, 10(2), 4-7.
Friese, G., Hendee, J. C., & Kinziger, M. (1998). The wilderness experience
program industry in the United States: Characteristics and dynamics.
Journal of Experiential Education, 21(1), 40-45.
Graziano, A., & Raulin, M. (1997). Research methods: A process of inquiry.
(3rd ed.). New York: Addison Wesley Longman.
Grotevant, H. D. & Cooper, C. R. (1986). Individuation in family relationships.
Human Development, 29, 82-100.
Hans, T. A. (2000). A meta-analysis of the effects of adventure programming
on locus of control. Journal of Contemporary Psychotherapy, 30(1),
33-60.
Hattie, J., Marsh, H. W., Neill, J. T., & Richards, G. E. (1997). Adventure
education and Outward Bound: Out-of-class experiences that make a
lasting difference. Review of Educational Research, 67(1), 43-87.
Jacobsen, N. S., & Truax, P. (1991). Clinical significance: A statistics approach
to defining meaningful change in psychotherapy research. Journal of
Consulting and Clinical Psychology, 59, 132-151.
44
Jenkins, M. (2000). The hard way. Outside, March, 45-52.
Kelly, F. J. (1974). Outward Bound and delinquency: A ten year experi-
ence. Paper presented at the International Conference of the Asso-
ciation of Experiential Education, Estes Park, CO.
Kelly, F., & Baer, D. (1968). Outward Bound: an alternative to institu-
tionalization for adolescent delinquent boys. Boston, MA: Fandel
Press.
Krakauer, J. (1995). Loving them to death. Outside, October, 1-15.
Lambert, M., & Cattani-Thompson, K. (1996). Current findings the effective-
ness of counseling: Implications for practice. Journal of Counseling
and Development, 74(July/August), 601-607.
Lambert, M., Ogles, B., & Masters, K. (1992). Choosing assessment de-
vices: An organizational and conceptual scheme. Journal of Coun-
seling and Development, 70(March/April), 527-532.
Lambert, M. J., Huefner, J. C., & Reisenger, C. W. (1996). Quality improve-
ment: Current research in outcome management. In W. T. Stricker &
S. Sheuman (Eds.), Handbook of quality management in behav-
ioral health.: Plenum.
Moote, G. T., & Wadarski, J. S. (1997). The acquisition of life skills through
adventure-based activities and programs: A review of the literature.
Adolescence, 32(125), 143-167.
Mosier, J. (1998). The predictive validity of the Youth Outcome Question-
naire: Prognostic Assessment. Unpublished dissertation, University
of Utah, Salt Lake City.
Mosier, J., Burlingame, G., Wells, G., Ferre, R., Latkowski, M., Johansen, J.,
Peterson, G., & Walton, E. (2001). In-home, family centered psychi-
atric treatment for high-risk children and youth. Children’s Services:
Social Policy, Research, and Practice, 4(2), 51-68.
Pitstick, R. (1995). Wilderness Discovery: The meaning and effects of a
seven-day wilderness experience program for youth-at-risk in the
federal Job Corps program (unpublished Doctoral dissertation).
Moscow: University of Idaho.
Robinson, K. (2000). Outcomes of a partial-day treatment program for
referred children. Child and Youth Care Forum, 29(2), 127-137.
Rogers, C. R. (1961). On becoming a person. Boston, MA: Houghton
Mifflin Company.
45
Russell, K. C. (2002). Perspectives on the wilderness therapy process and its
relation to outcome. Child and Youth Care Forum, In Press.
Russell, K. C. (2001). Assessment of treatment outcomes in Outdoor Behav-
ioral Healthcare. (Technical Report 27). Idaho Forest, Wildlife, and
Range Experiment Station, Moscow, ID: Available from the UI-Wilder-
ness Research Center-Outdoor Behavioral Healthcare Research Coop-
erative (http://www.its.uidaho.edu/wrc).
Russell, K., & Hendee, J. (2000). Outdoor behavioral healthcare: Defini-
tions, common practice, expected outcomes and a national survey
of programs (Technical Report 26). Idaho Forest, Wildlife, and Range
Experiment Station, Moscow, ID: Available from the UI-Wilderness
Research Center-Outdoor Behavioral Healthcare Research Cooperative
(http://www.its.uidaho.edu/wrc).
Russell, K. C. (2000). Exploring how the wilderness therapy process relates to
outcomes. Journal of Experiential Education, 23(3), 170-176.
Russell, K. C. (1999). Theoretical basis, process, and reported outcomes of
wilderness therapy as an intervention and treatment for problem
behavior in adolescents. Unpublished doctoral dissertation, University
of Idaho, Moscow, ID.
Stinchfield, R. D. N., L.; Feder, S. (1994). Follow-up contact bias in adoles-
cent substance abuse treatment outcome research. Journal of Studies
on Alcohol, 55, 285-289.
Vermillion, J., & Pfeiffer, S. (1993). Treatment outcomes and continuous quality
improvement: Two aspects of program evaluation. Psychiatric Hospi-
tal, 24, 9-14.
Wells, M. G., Burlingame, G. M., Lambert, M. J., Hoag, M. J., & Hope, C. A.
(1996). Conceptualization and measurement of patient change during
psychotherapy: Development of the Outcome Questionnaire and Youth
Outcome Questionnaire. Psychotherapy, 33(Summer), 275-283.
Wells, S. E. (1990). At-risk youth: Identification, programs, and recommen-
dations. Engelwood, CO: Teacher Ideas Press.
Winterdyk, J., & Griffiths, C. (1984). Wilderness experience programs: reform-
ing delinquents or beating around the bush? Juvenile and Family Court
Journal, Fall, 35-44.
Winters, K. C. (1999). Treating adolescents with substance use disorders: An
overview of practice issues and outcomes. Substance Abuse, 20(4),
203-225.
... Wilderness treatment is one option of care that effectively treats children and adolescents presenting aggressive behaviour [28]. Wilderness therapy combines traditional therapy techniques with group therapy in a wilderness setting, approached with therapeutic intent [29]. Adolescents demonstrated marked improvements in the following areas: anxiety and depression, substance abuse and dependency, disruptive behaviour, defiance and opposition, impulsivity, suicidality, violence, sleep disruption, school performance, and interpersonal relationships. ...
... Adolescents demonstrated marked improvements in the following areas: anxiety and depression, substance abuse and dependency, disruptive behaviour, defiance and opposition, impulsivity, suicidality, violence, sleep disruption, school performance, and interpersonal relationships. Russel [29] conducted a study lasting for 45 days. Adolescent client well-being was evaluated using the Youth Outcome Questionnaire (Y-OQ) and the Self Report-Youth Outcome Questionnaire (SR Y-OQ) [30]. ...
... Although the idea of OBH and wilderness therapies is remarkable and the study by Russell [29][30][31] yields positive results, he admits that OBH and wilderness therapy's effectiveness reveals a consistent lack of theoretical basis, methodological shortcomings and problematic results difficult to replicate [31]. Therefore, we have focused on the primary tool for influencing attitudes and behaviour: remedial educative programmes [52] based on Shinrin-yoku, observational learning and forest pedagogy methods. ...
Article
Full-text available
This paper evaluates the impact of the forest environment on aggressive manifestations in adolescents. A remedial educative programme was performed with 68 teenagers from institutions with substitute social care with diagnoses F 30.0 (affective disorders) and F 91.0 (family-related behavioural disorders), aged 12–16 years. Adolescents observed patterns of prosocial behaviour in forest animals (wolves, wild boars, deer, bees, ants, squirrels and birds), based on the fact that processes and interactions in nature are analogous to proceedings and bonds in human society. The methodology is based on qualitative and quantitative research. Projective tests (Rorschach Test, Hand Test, Thematic Apperception Test) were used as a diagnostic tool for aggressive manifestations before and after forest therapies based on Shinrin-yoku, wilderness therapy, observational learning and forest pedagogy. Probands underwent 16 therapies lasting for two hours each. The experimental intervention has a statistically significant effect on the decreased final values relating to psychopathology, irritability, restlessness, emotional instability, egocentrism, relativity, and negativism. Forest animals demonstrated to these adolescents ways of communication, cooperation, adaptability, and care for others, i.e., characteristics without which no community can work.
... Working with adolescents with emotional and behavioural problems such as ADHD in a natural environment has been reported to be a common practice in the West (Davis-Berman et al., 1994). It has been reported that integrating individual and group therapy with challenging tasks promoted positive outcomes in adolescents such as reducing distress (Russell, 2003). A Delphi study by Burg (1994) suggested that family group psychotherapy conducted in the natural outdoors seems to be beneficial to the development of family strengths, and pinpointed the need for more process research on the use of natural environments in this intervention modality. ...
Article
This article reports the therapeutic resources in the natural outdoors that were conducive to the therapeutic process in a multi‐family therapy (MFT) for enhancing the self‐efficacy and collective family efficacy of Chinese families of adolescents with attention deficit hyperactivity disorder (ADHD) in Hong Kong. Photo‐elicited interviews with adolescents with ADHD and parent focus groups were employed to explore the participatory experiences of the families. A total of thirteen Chinese families of adolescents (aged 11–15 years) with ADHD participated in this study (fourteen adolescents with ADHD, eight fathers and ten mothers). The present study explored the potential therapeutic resources in the natural outdoors, namely a change in the group environment, spaciousness and darkness. The use of the natural outdoors in MFT proved to be a strategy useful for creating a naturalistic group setting within which family participants can more easily enact changes.
... It is well-documented that for young people, physical activity combined with nature contributes a range of significant social benefits, including intra-and interpersonal development, crime reduction and active citizenship (Eigenschenk et al. 2019). This is particularly important for at-risk youth, for which these adventure activities bring a number of benefits, such as raising self-esteem, enhancing dependency recovery and reducing symptoms of behavioural issues (Russell 2003). Older individuals benefit from opportunities for euphoria through exercise. ...
Article
Non-motorized water sports requiring physical efforts such as swimming, scuba diving, kayaking, sailing and surfing are becoming increasingly popular in Mediterranean marine protected areas (MPAs). This research investigates the relationship between these types of water sports and practitioners’ psychological and mental health. It takes the MPAs of Cap de Creus and Gulf of Roses (north-western Mediterranean) as a case study and is underpinned by a literature review and in-depth interviews with specialized water sports instructors. Results provide evidence that doing non-motorized water sports in the sea has positive outcomes for practitioners’ physical and mental health. When done in well-preserved areas, these sports may be a viable tool for both wellness and health recovery, and could be introduced in the community as a preventative and rehabilitation health strategy. This should be accompanied by strategies to address the ecological impacts these sports may have on MPAs.
... Also in outdoor therapies, the focus has mostly been on the inner subjective experiences of the individual rather than interpersonal relationships or structural challenges (Harper et al. 2007). Research evaluating various versions of outdoor therapies have largely been concerned with measuring significant symptom reductions (Clark et al. 2004;Russell 2003), which may be argued in line with the clinical understanding of recovery as result-oriented and symptom-focused (Desai 2018;Macpherson et al. 2016). However, the diagnostic focus runs the risk of missing the fact that although the person may have developed, the situation or context to which they are going back might not have changed (Desai 2018). ...
Chapter
In line with the increasing mental health challenges across the world, attention has been called to the human–nature bond and the possible detrimental consequences of dislocation and alienation from nature. Following this attention, there has been a rapidly growing interest in the potential health and welfare benefits of a nurturing relationship with nature within research, politics, and practice. This entry critically explores perspectives on recovery and well-being related to interactions between humans and the more than- human-nature. Recognizing that health is largely maintained through daily life, focus is dedicated to engagement with nature as promoting well-being through experiences that are supportive in living and dealing with difficult life situations. A socio-ecological approach to persons’ mental health recovery is argued as important for addressing the interrelationship between the increasing inequalities of individuals’ mental health worldwide and the global ecological situation.
Article
Nature-based interventions hold promise for vulnerable youth experiencing mental, emotional, developmental, behavioral, or social difficulties. This scoping review examined wilderness therapy, animal assisted therapy, care farming, and gardening and horticultural therapy programs to raise awareness and guide future development of research and treatment options. Studies included in this review were identified through a systematic search of the literature informed by a scoping review framework. Studies were examined by design, sample, intervention, and key findings. The majority of studies were quantitative using repeated measures designs and were conducted primarily in the United States. Sample sizes were generally small. Interventions were residential and community based with varying degrees of duration. Outcomes were largely positive across a wide range of psychosocial and behavioral measures and often maintained post-treatment. We emphasize the importance of robust empirical designs, comprehensive description of the interventions and surrounding therapies, and identification of target groups.
Article
Context: Daily outdoor play is encouraged by the American Academy of Pediatrics. Existing evidence is unclear on the independent effect of nature exposures on child health. Objective: We systematically evaluated evidence regarding the relationship between nature contact and children's health. Data sources: The database search was conducted by using PubMed, Cumulative Index to Nursing and Allied Health Literature, PsychInfo, ERIC, Scopus, and Web of Science in February 2021. Study selection: We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. In all searches, the first element included nature terms; the second included child health outcome terms. Data extraction: Of the 10 940 studies identified, 296 were included. Study quality and risk of bias were assessed. Results: The strongest evidence for type of nature exposure was residential green space studies (n = 147, 50%). The strongest evidence for the beneficial health effects of nature was for physical activity (n = 108, 32%) and cognitive, behavioral, or mental health (n = 85, 25%). Physical activity was objectively measured in 55% of studies, and 41% of the cognitive, behavioral, or mental health studies were experimental in design. Limitations: Types of nature exposures and health outcomes and behaviors were heterogenous. Risk of selection bias was moderate to high for all studies. Most studies were cross-sectional (n = 204, 69%), limiting our ability to assess causality. Conclusions: Current literature supports a positive relationship between nature contact and children's health, especially for physical activity and mental health, both public health priorities. The evidence supports pediatricians in advocating for equitable nature contact for children in places where they live, play, and learn.
Article
This paper reviews three psychoanalytic conceptualizations of substance misuse, discussing the implications for substance treatment deriving from these conceptualizations. The paper also presents wilderness therapy as an evidence-based approach, reviewing the outcomes and primary components of such programs. Finally, the paper concludes with two case studies of adolescents with substance use disorders treated in wilderness therapy.
Article
Objective The purpose of this study was (1) to examine whether three domains of a parent's functioning—parent distress, interpersonal relationships, and social role performance—changed over the course of their child's treatment, (2) to examine how these factors as measured at intake predicted youth progress in psychotherapy, and (3) to examine whether changes in these parent factors over the course of youth psychotherapy were associated with changes in youth symptoms. Method Participants were 339 youth, ages 4–17 and their parents from a community outpatient treatment setting undergoing usual care. Parent and child outcomes were examined across five time points over the course of child treatment. Hierarchical linear modeling was used to examine the relationship between parent domains and youth progress in therapy. Results Results suggested that parent domains significantly predicted their child's symptoms at intake as well as change in psychotherapy. In addition, parent domains improved over the course of youth treatment and the progression of these changes was related to the progression of changes in youth scores across the course of treatment. Conclusions The results of this study highlight the important relationship between youth and parent functioning in the context of treatment of youth mental health issues.
Article
Full-text available
Adventure practitioners asked to justify their work with adolescent populations have no one study to point to that statistically sums up major findings in the field. Whether it be a school board, treatment facility, or funding agency, one study is needed which can combine statistics from many studies into a format to show overall effectiveness of adventure programming. This study used the statistical technique of meta-analysis to demonstrate that adolescents who attend adventure programming are 62% better off than those who do not. While combining various populations and outcomes resulted in an overall effect that could be considered small by some accounts, the study did point to major problems with current research and offers some direction for future researchers to explore.
Article
Full-text available
Thirty one therapeutic wilderness programs specializing in mental health treatment were examined through both a mail survey and a telephone interview. Respondents provided information on sponsorship, type of clients served, outdoor modality, therapeutic modality, staff qualifications, and other professional issues. The results suggested that most programs were serving high-risk adolescents using a variety of outdoor modalities. Programs had a difficult time describing their therapeutic process, and there was a lack of research and follow-up in the programs. The lack of basic definitions of therapeutic, non-professional staff, and the lack of process and outcome research are discussed. Finally, the authors call for dialogue on these and other professional issues related to therapeutic wilderness programs.
Article
Full-text available
The Nokomis Challenge Program, an innovative correctional program for low- and medium-risk delinquents, was implemented by the Michigan Department of Social Services (DSS) in 1989. The program combines three months of residence and outdoor challenge programming with nine months of intensive community-based aftercare. It was designed as an alternative to residential placements that average 14-16 months. An evaluation of the Nokomis Program conducted using a quasi-experimental study of 192 juvenile offenders showed significant cost savings, compared to traditional residential placement, over a 24-month period, but few differences in outcomes. Youths in both groups made positive gains in social adjustment during residential programs, which disappeared by the end of the follow-up period. Despite community aftercare, Nokomis youths failed more quickly upon release to the community. The results suggest that to derive benefit from short-term alternative placements, the intensive aftercare component must be strengthened to help prevent youth from relapsing.
Article
Outdoor behavioral healthcare (OBH) is an emerging intervention and treatment method in mental health practice which helps adolescents overcome emotional, adjustment, addiction, and psychological problems. OBH programs apply wilderness therapy as a treatment approach, which involves immersion in an unfamiliar environment, group living with peers, individual and group therapy under the supervision of licensed professionals, and educational curricula. The goal of OBH treatment is to reveal and address problem behaviors, foster personal and social responsibility, and enhance the emotional growth of adolescent clients. A nation-wide survey of 86 OBH programs that annually serve over 10,000 clients and their families showed that over 80% of all OBH programs are licensed by state agencies. Results also showed that there are more private placement than adjudicated programs by nearly five to one, and that less than 25% of client treatment costs are covered by insurance payment. OBH program models vary and are often used as an alternative treatment for adolescents not reached by traditional counseling services.
Article
Despite a growing number of programs operating in the United States under the guise of “wilderness therapy,” a consistent and accepted definition is lacking. Rehabilitative, outdoor-based approaches, such as “challenge courses,” “adventure-based therapy,” or “wilderness experience programs,” are often used interchangeably to describe “wilderness therapy.” This paper attempts to shed light on this issue by presenting an integrated and consistent definition of wilderness therapy to guide program design and future research efforts. Three trends in the industry are also explored which support the notion that wilderness therapy programs are searching for recognition by state agencies, national accreditation agencies, insurance companies and mental health professionals. These trends support the idea that consistent wilderness therapy practices may be emerging which support the use of a consistent definition.
Article
Wilderness therapy is seen as a treatment option for seriously troubled adolescents not being reached by traditional forms of treatment. The research shows that wilderness therapy can improve self-perceptions, increase social adjustment, and reduce recidivism of adolescent participants. However, research on wilderness therapy has not been specific in describing how presenting problems are assessed by wilderness therapy and how therapeutic approaches relate to target outcomes. This article examines the wilderness therapy process in context to illustrate how the process related to specific outcomes for four client case studies in four wilderness therapy programs. Trends emerged from qualitative data analysis illustrating the importance of alone time and opportunity for reflection and a non-confrontive and caring approach that helped clients establish a trusting relationship with staff. Three outcomes were common across all four client case studies in the form of proposed changes: (a) a better relationship with their family, (b) abstinence from drugs and alcohol, and (c) a desire to try harder and complete school. Four-month follow-up interviews revealed clients doing well, though three had used drugs and alcohol to varying degrees, which underlines the importance of structured aftercare services.